The goal of this project is to understand the impact of marijuana-related legislation on adolescent marijuana and other substance use and related risk factors using time-series analysis of survey data from Colorado, Washington, and Oregon and two additional comparison states. This R21 (exploratory/developmental) study will examine changes in the age of onset, and prevalence of any use, and regular use of marijuana among youth in these states, and lay a foundation for a future nationwide examination of the impact of state-level marijuana legislation on marijuana-related risk factors and outcomes among youth. The recent legalization of recreational marijuana use for those over 21 years of age in Washington and Colorado and the failure of legalization in Oregon, along with the medical marijuana laws of earlier decades, have changed the legal and normative context for current and future generations of youth. One of the primary concerns for public health is that legalization may lead to more favorable attitudes and norms and lower perceived harm from marijuana use, which have been shown to predict earlier age of marijuana use onset and greater adolescent use, which, in turn, have been associated with a range of negative outcomes, including subsequent drug abuse and dependence and interference with a healthy and successful transition into adulthood. As the national discussion around legalization of marijuana proceeds, there is an urgent need for scientific evidence from timely, carefully implemented studies informing whether such changes in marijuana laws affect youth development. The proposed study will use an interrupted time-series design (ITS), including multiple baseline ITS, to examine statewide survey data from adolescents and thereby assess the long-term patterns of age of onset and youth marijuana use and marijuana-related attitudes, norms, and perceived harm of using marijuana among youth before and after (1) medical marijuana legalization (MML; passed in 1998 in WA and OR, and 2000 in CO), and (2) recreational marijuana legalization (RML; passed in 2012 in WA and CO, and was rejected by the voters in the same year in Oregon), with additional comparisons to state-wide survey data from Montana and Massachusetts (which enacted MML in 2004 and 2012, respectively; and neither had RML on the ballot). Moreover, it is possible that these legislative changes affect other adolescent substance use. Thus, this study will examine the patterns of change in youth alcohol, tobacco, and other drug use in these three states over time to assess the plausibility of the so-called substitution effect in which marijuana use substitutes (or, in contrast, exacerbates) other substance use. The data for the analytic part of the proposed R21 study will come from the existing and planned statewide student surveys from Grades 6-12 in Colorado, Washington, Oregon, and additional comparison states of Montana and Massachusetts.